Tomato Severity Segmentation Model API

A FastAPI-based application for detecting and analyzing disease severity in tomato plants using image segmentation techniques.

Overview

This API allows users to upload tomato plant images and analyze them for various diseases and their severity. The system uses a segmentation model to identify affected areas and classify the severity of diseases such as early blight, late blight, bacterial spot, and septoria leaf spot.

Features

• Image Upload: Upload tomato plant images for analysis
• Disease Detection: Identify multiple disease types in a single image
• Severity Classification: Classify the severity of detected diseases
• Segmentation Maps: Generate segmentation maps highlighting affected areas
• Health Monitoring: Built-in health endpoint for monitoring application status

Installation

Prerequisites

• Python 3.8+
• pip (Python package manager)

Setup

1. Clone the repository:

   git clone <repository-url>
   cd tomatoseveritysegmentationmodel
   

2. Install dependencies:

   pip install -r requirements.txt
   

3. Run database migrations:

   alembic upgrade head
   

4. Start the application:

   uvicorn main:app --reload
   

The API will be available at http://localhost:8000

API Endpoints

Health Check

• GET /health: Check the health of the application

Image Management

• POST /api/tomatoes/upload: Upload a tomato image
• GET /api/tomatoes: List all uploaded tomato images
• GET /api/tomatoes/{image_id}: Get details of a specific image
• DELETE /api/tomatoes/{image_id}: Delete an image and its analysis data

Analysis

• POST /api/model/analyze/{image_id}: Analyze a tomato image for disease severity
• GET /api/model/results/{image_id}: Get all analysis results for a specific image
• GET /api/model/info: Get information about the segmentation model

Database Schema

The application uses SQLite with the following main tables:

• tomato_images: Stores uploaded image metadata
• analysis_results: Stores analysis results for each image
• severity_details: Stores detailed severity data for each analysis

Segmentation Model

The current implementation uses conventional computer vision techniques for segmentation as a proof of concept. In a production environment, this would be replaced with a trained deep learning model like U-Net or DeepLabV3.

The model classifies the following disease categories:

• Healthy
• Early Blight
• Late Blight
• Bacterial Spot
• Septoria Leaf Spot

Example Usage

Upload an image

curl -X POST "http://localhost:8000/api/tomatoes/upload" \
  -H "accept: application/json" \
  -H "Content-Type: multipart/form-data" \
  -F "file=@tomato_plant.jpg"

Analyze the image

curl -X POST "http://localhost:8000/api/model/analyze/{image_id}" \
  -H "accept: application/json"

Development

Project Structure

tomatoseveritysegmentationmodel/
├── app/
│   ├── api/
│   │   └── routes/
│   ├── core/
│   ├── db/
│   ├── models/
│   ├── schemas/
│   ├── services/
│   └── utils/
├── migrations/
├── storage/
│   ├── db/
│   ├── images/
│   └── models/
├── alembic.ini
├── main.py
└── requirements.txt

Adding New Features

To add new features:

1. Define models in app/models/
2. Create Pydantic schemas in app/schemas/
3. Add database migrations with Alembic
4. Implement routes in app/api/routes/
5. Update tests

License

This project is licensed under the MIT License - see the LICENSE file for details.